The tropical cyclone (TC) size, which is generally defined as the azimuthally-averaged radius of the ocean-surface winds, has been widely studied in the recent decades. However, as a matter of fact, TC size is often asymmetric. Understanding the asymmetry of TC size is particularly important for the coastal regions, which receives limited discussion. In this conference, we will show the 40-yr global climatology (1979-2018) of TC size asymmetry using the ERA5 reanalysis data. Both the temporal and spatial characteristics of the TC size asymmetry will be discussed.

When there is a tropical cyclone recurving from the western North Pacific to the east of Luzon Island of Philippines, its topography could induce tropical cyclogenesis over South China Sea. There were two extraordinary cases in 2019 and 2020. As Tropical Storm Jangmi (2020) was skirting the east of Luzon Island, a tropical depression Mekkhala (2020) started to form to the west of the island, which seems similar to the formation of tropical disturbance 91W in 2019. Tropical depression Mekkhala (2020) soon intensified to typhoon and made landfall on South China. In this study, the evolution of Typhoon Mekkhala (2020) is simulated and terrain sensitivity experiment is conducted using the WRF model. Possible mechanisms of the tropical cyclogenesis of Mekkhala (2020) are discussed based on the results of numerical experiments.

Typhoon Chaba (2016) made landfall on the Korean peninsula and caused record-breaking rainfall in southeastern Korea. In particular, Ulsan metropolitan region experienced a 300-year flood due to 319 mm of rain for 3 hours. The reasons for heavy rainfall induced by typhoon Chaba were possibly associated with mountainous topography in southeastern Korea and warm sea surface temperature (SST) anomaly in the East China Sea. In this study, we investigated the effects of topography and SST anomaly through numerical experiments with high-resolution Weather Research and Forecasting (WRF) model. To verify the topographic effect in southeastern Korea, we set 3 nested domains with horizontal grid spacings of 9-3-1km and conducted the control experiment (CTL) and the reduced topography experiment (TOPO). As a result, the CTL experiment simulated more rainfall on the windward side due to the lifting effect by mountain compared to the TOPO experiment. The air flow in the CTL experiment was mechanically lifted by mountain slopes, and the orographic lifting enhanced precipitation on the windward side. We also conducted the experiment with climatology sea surface temperature in the East China Sea (CSST) to examine the effect of SST anomaly on typhoon-induced precipitation. The CSST experiment simulated less heavy rainfall compared with the CTL experiment because cold climatological SST in the East China Sea led to decreased TC intensity and at landfall time. Therefore, the numerical experiments indicated that heavy rainfall in Korea induced by typhoon Chaba could be enhanced by the effects of intrinsic topography in southeastern Korea as well as warm SST anomaly in the East China Sea.

The genesis of tropical cyclones (TCs) over the western North Pacific (WNP) mainly occurs during June-October, and TCs result in significant damages in East Asian countries (e.g., Korea, Japan, Taiwan, and China, etc.). In August and September 2019, TCs occurred similar to normal, but the numbers of TCs affecting Korea were three, respectively, that recorded above normal (1982-2018). Over the analyzed 38 years (1982-2019), it was only two months (August 2004 and July 2014) that affected three typhoons in Korea except for 2019. In this study, therefore, we examine the cause of the largest TCs migrated into Korea in 2019. In July 2019, the number of TCs affecting Korea was normal compared to those in August and September. Thus, we analyzed environmental conditions of three months, which can influence TC activities (e.g., steering flow, geopotential height at 500 hPa, vertical wind shear (VWS), and sea surface temperature (SST)). As a result, the tracks of TCs affecting in Korea were significantly associated with steering flow during July-September. Furthermore, weaker VWS and higher SST were distributed in nearby TC tracks affecting Korea during August and September, whereas strong VWS and lower SST were dominant in July. These environmental conditions in August and September were favorable for maintaining and developing TCs, which can explain why the largest typhoons affected Korea more than normal during August and September in 2019.

Classification of precipitation types is one of the most challenging research fields in atmospheric sciences. The surface temperature, surface wet bulb temperature, thicknesses between 1000–850 mb, and so on have been used to classify the precipitation types. In addition to these variables, the recent dual-polarization radar network in Korea enables accurate precipitation type classification with high resolution. However, a complicated algorithm is required to incorporate these useful variables into characterizing precipitation type. Machine learning (ML) techniques are promising tools to solve this issue because it appropriately deals with “big data” and complex relationships between a response and predictors without any distributional assumption. In this study, we have examined the capability of three supervised ML methods (decision tree, random forest, and support vector machine) in classification of surface precipitation types for 2018–2019 winter events. The ML methods were trained by manual surface measurement data (no precipitation, rain, mix, and snow) as a multinomial response, and three-dimensional dual-polarization variables and NWP analysis field as predictors. We conducted a predictor selection based on the importance of variables from the random forest and optimized the ML methods. The hold-out evaluation was performed to compare the accuracy of ML methods. In the order of random forest (0.897), support vector machine (0.861), and decision tree (0.789), the prediction accuracy of classification was higher. For the potential application to the real-time classification, we preliminary applied the best ML model to the operational polarimetric radar network. ACKNOWLEDGEMENT: This research is supported by "Development and application of Cross governmental dual-pol. radar harmonization (WRC-2013-A-1)" project of the Weather Radar Center, Korea Meteorological Administration. This work was funded by the Korea Meteorological Administration Research and Development Program under Grant KMI2020-00910. 

Precipitation systems formed in convective and stratiform region had different structure caused by different growing mechanisms. Since the high impact weather such as flash flood, lightning, and hail occurred in convective storm with strong updraft, accurate discrimination of precipitation types is important. The dense radar network operated by KMA (Korea Meteorological Administration) can provide 3 dimensional dual-polarization variables (e.g. Z_h, Z_DR, K_DP, ρ_HV) with high spatiotemporal resolution. Therefore, we develop the classification algorithm of precipitation types based on 3D radar data in Korean peninsula.  Classification procedure is performed using two algorithms consisting of SL3D (storm labeling in 3 dimension) and categorization method based on feature parameters. The areas of convective precipitation are subdivided into convective and updraft, while those of stratiform segmented into precipitation and non-precipitation stratiform. The deep system zone can be appeared from both convective and stratiform areas, and shallow system and anvil cloud are detected in this algorithm. The validation is carried out to confirm adequateness of classified precipitation types and physical property of rainfall vertical structure is investigated using radar-based observations.  The validation results shows that classification algorithm is more reasonable than one of previous study in respect to vertical structure of precipitation systems. The peaks of vertical reflectivity profiles existed at 5 km in stratiform cloud, while those of K_DP indicated at 4.5 km in deep system. In addition, the maximum (minimum) values of vertical profiles for Z_DR (ρHV) formed at 5.0 (~5.5) km height in both stratiform and deep system cloud. Consequently, the bright band formed at 4.5~5.5 km during warm seasons in Korea. The mean values of vertical reflectivity profiles for convective precipitation are higher than those of stratiform at all atmospheric layers. ACKNOWLEDGEMENT: This work was funded by the Korea Meteorological Administration Research and Development program under Grant KMI2020-00910. 

In this study, we analyze the surface PM2.5 concentrations over Taiwan observed under cold-season weak synoptic weather, focusing on how the local circulation with complex topography modulates the spatial distribution of the particulate pollutants, as well as the exacerbation of PM2.5 pollution by the duration of consecutive weak synoptic conditions. Weak synoptic weather condition in winter and spring is most frequently associated with the heavy PM2.5 pollution events in Taiwan, when the formation of lee-side vortices and/or weak areas on the west side of Taiwan influences the dispersion of locally-emitted pollutants. However, in addition to the local circulation, the observed PM2.5 concentrations under weak synoptic regime could be influenced by the distance to local emission sources, vertical mixing, and chemical reactions. The goal of the present study is to delineate the observed signal of surface PM2.5 predominantly contributed by local circulation. We selected 302 weak synoptic weather days in spring and winter during 2008-2019 based on the Taiwan weather event list, rainfall stations data, and the sounding profiles at Ishigaki Island. We then defined the PM2.5 ratio, which is the daily PM2.5 concentration divided by annual mean PM2.5 at each Taiwan Environmental Protection Agency (EPA) station. Under weak south-easterly synoptic wind, the stations with high-frequency of enhanced PM2.5 are located at the lee-side of Central Mountain Range consistent with the trapping of mountain blocking effect. Moreover, pollution enhancement evolves into a more serious condition in the consecutive weak synoptic days, and the high-frequency areas extend northward to Taipei basin on the day. The enhanced pollution areas identified by the PM2.5 ratio are more consistent with the weak wind region than the composite mean value of PM2.5 concentration. The current results provide valuable evaluation metrics for the dispersion of near-surface air pollutants by local circulation in high-resolution numerical models. 

Downslope windstorms occur frequently in spring and winter in the lee of the Eastern Mountains (Yeongdong region) in Korea. Generation mechanisms are dependent on background flows revealed in synoptic-scale patterns. We classified synoptic conditions for downslope windstorm events using Self-Organizing Map (SOM). First, we define the events when daily maximum wind speed exceeds 20 m s^-1 at one of two stations in Yeongdong region for 41 years (1979-2019). For these events, sea level pressure (SLP) anomaly from the NCEP-DOE reanalysis 2 data is used as an input for SOM. Rapidly increasing Explained Cluster Variance (ECV) is saturated when the number of clusters is 8. Among 8 clusters, four synoptic SLP patterns are identified: 1) south high and north low in spring, 2) west high and east low in winter, 3) strong low in East Sea and 4) transition between 1 and 2. Those patterns are statistically significant with 99% confidence level. 850-hPa synoptic fields show that warm (cold) advection along with southwesterly (northwesterly) is in the type 1 (2). Surface temperature reveals that warming effect due to downdraft is the strongest in the type 1, which is likely related to severe wildfires in this area. The airflows examined using ERA5 data show that the mechanisms of downslope windstorms are different in each pattern. More events in the type 2 are explained by hydraulic jump theory due to frequent inversion layers and upstream flows with moderate Froude number. According to vertical wave number profiles, partial reflections of waves occur more often in springtime patterns. The cases with wave breaking at critical level are mostly included in the type 1. Acknowledgement: This research was supported by the Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education (NRF-2019R1I1A2A01060035). 

Much of the water vapor flowing inland from the oceans is converted to precipitation in coastal areas due to the orographic effects of the coastal mountain ranges. In the tropics, the characteristics of precipitation concentrated around the coastlines has been investigated by using the Tropical Rainfall Measuring Mission (TRMM) Precipitation Radar (PR), which can observe precipitation regardless of surface condition. Although the TRMM PR observations were limited to low-latitude regions, the accumulation of observation data by the Dual-frequency Precipitation Radar (DPR) onboard the Global Precipitation Measurement (GPM) Core Observatory launched in 2014 has made it possible to obtain the distribution of coastal precipitation at higher latitudes. To reveal how low-level wind speed controls the spatiotemporal distributions of precipitation around the coastlines, the precipitation datasets from the two spaceborne radars are presented as a function of the distance from the coastlines and the landward low-level wind speed. Under conditions of strong landward winds in the tropics, low-level prevailing winds, such as monsoon flows, are forced to rise on the windward side of the coastal mountain ranges, bringing continuous precipitation with small diurnal variation to coastal areas. When the wind is weak in the tropics, the diurnal variation of precipitation caused by the heating contrast between land and sea is dominant, and the propagations of precipitation away from the coastline are identified in both land and ocean. As the latitude increases, the amount of solar radiation incident on the earth's surface gets smaller, thus the diurnal variability of precipitation gets smaller, and large amounts of coastal precipitation occur under strong landward wind conditions.

Typhoons are a disturbance with an overall scale of 1000km while meso-scale systems that form within them are a phenomenon with a scale of only a few 10km. When Typhoon Trami (2018) passed Okinawa Island on September 29-30, weather radars on Okinawa Island observed the inner core region of the typhoon. The purpose of this study is to reveal characteristics and structure of the meso-scale structure of wind and thermodynamic fields that occurred in the eye of Typhoon Trami. In this study, we used Meteorological Satellite Himawari-8, JMA RSMC best-track data, surface observation, NICT Okinawa Bistatic Polarimetric RAdar, and Phased Array Weather Radar. The radar observation showed that the size of Trami eye was larger than 100km in radius, which was a very large and polygonal eye. A strong wind was observed with sharp drop of the equivalent potential temperature during the passage of the eye and the inner edge of the eyewall. The decrease of equivalent potential temperature lasted for about 40 minutes and was accompanied by an increase of wind speed. The mixing ratio also decreased during the period. This indicates that a dry air passed over the observation sites and that it was accompanied by the strong wind. To investigate the vertical motion in the inner core region, we developed a CVAD method and used it to analyze the wind field. The CVAD method derives the vertical distribution of horizontal velocity above the radar from Doppler velocity data obtained by PAWR. This analysis shows that the downdraft was pronounced during the period when the equivalent potential temperature was decreasing. In this study, we identified the meso-scale structure causing strong winds that occur at the surface in the eye of Typhoon Trami. This strong wind was observed when the inner edge of the polygonal eyewall approached. 

Formation and developing processes of hydrometeors are important to understand rainfall systems. To observe cloud and precipitation particles, hydrometeor videosonde (HYVIS) was developed and improved recently. However, manual analysis of a large amount of video images takes time to identify particle type and difficult to classify the particles objectively. It is required to develop automatic particle identifying system. Purpose of this study is to develop particle detection and identification system using deep learning and to clarify the micro-physical characteristics of subtropical rainfall systems. We prepared training, verifying and testing image data in the region above 0℃ level observed in Palau Islands in 2013. Numbers of sheets of each data are 508, 100 and 6714. Annotations (Labeling) of training and verifying data are classified into eight categories: column (CO), column plate (CP), supercooled water (SW), plate(PL), aggregates (AG), frozen drop(FD), undetermined(UD), and plate-broken(PD). First, we took 30 frames of HYVIS image per a second. Next, in order to make particles to identify more clearly, all of snapshots were transferred to gray scale, decreased brightness and enhanced contrast. Then, clear shaped pictures were chosen from those pictures as training and verifying data and they were labeled as one of the eight categories. Labeled data were studied by a software named yolov5, which is one of the deep learning algorithms .By comparing results of studied and verifying data labels, we checked the accuracy of studied data. Finally, verified-studied data were used to examine test data. CO and SW were identified as good as manual identification, but others were wrong identified or undetected by the algorism. We infer that these wrong results were caused by the lack and deviation of training data. Uncertainness of labeling might be one of the reasons for the failed identification. 

Using global precipitation datasets (GSMaP, TRMM) and the latest reanalysis data (ERA5) we performed a comprehensive analysis of the tropical rainfall variability that accompanies global-scale, low-frequency normal modes: Rossby, Rossby-gravity and Kelvin modes. Cross spectral analysis and lag-regression analysis both showed that coherent rainfall variations accompany not only the wavenumber 1 gravest Rossby mode (“5 day” wave) but other low-frequency modes. The normal mode rainfall variations are enhanced in regions such as the Amazon basin, but also include circumglobally travelling structures with substantial amplitude over the open ocean. These results are remarkably consistent among the three datasets including even ERA5 rainfall data. The circumglobal rainfall signals may be considered primarily as a response to the normal mode dynamical variations. We found that the phase relationship between rainfall and dynamical field variability is strongly dependent on the type of mode and even on the zonal wavenumber. We suggest that this is explained by the difference in relative importance of two underlying processes: (1) moisture-flux convergence and (2) rainfall enhancement associated with adiabatic cooling. Our determined rainfall signals are the response to quasi-monochromatic, periodic waves that have a simple vertical structure and represent one special case of tropospheric wave-rainfall coupling. Implications for the mechanism of 12-hr rainfall oscillations believed to be forced by the atmospheric tide are also considered.

Heatwaves cause abnormally high surface temperatures, damage human health, and cause social and economic losses. This study classified the heatwaves in the Korean Peninsula into three types by performing an empirical orthogonal function analysis of surface temperature on hot summer days during 1991–2020. Typical heatwave days are those with low wind speed and high pressure (HP); those with high pressure in the north, low pressure in the south, and easterly wind (EW); and those with both high pressure and easterly wind (HPEW). The average surface temperature in the Korean Peninsula was the highest on HPEW days, followed by HP and EW days. HPEW and EW days had higher surface temperatures in the western region of the Korean Peninsula than in the eastern region owing to the Foehn effect caused by the mountain ranges located in the north–south direction in the Korean Peninsula; however, HP days were not the cause of the low wind speed. On HPEW and EW days, the sea surface temperature in the East Sea (Sea of ​​Japan) and the surface temperature in the Korean Peninsula had a positive correlation, whereas the relationship between the two was not statistically significant on HP days. Using a high-resolution model, experiments were performed to increase the sea surface temperature in the East Sea, and the results were confirmed. The results of this study suggest that the influence of the ocean should be treated as an important factor when studying heatwaves in the Korean Peninsula. This work was funded by the Korea Meteorological Administration Research and Development Program under Grant KMI2020-01212.

The positive barotropic structure of the geopotential height (GPH) anomaly over the Ural Mountains has been steadily increasing since the late 1990s, and its strengthening has played an important role in the development of extreme heat waves and droughts in Western Europe and eastern Russia during that period. In this study, the underlying mechanism of this strengthening is investigated through analysis of the second empirical orthogonal function (EOF) mode of the 850 hPa GPH anomaly over the Eurasian continent, which is broadly characterized by positive anomalies over the Ural Mountains. Our results suggest that the second leading mode is closely correlated with the decadal variation in the Silk Road pattern (SRP) phase transition from positive to negative since the late 1990s (r = –0.71). The negative phase of the SRP is accompanied by a southerly (northerly) wind anomaly with warm (cold) advection over the western (eastern) boundary of the positive GPH anomaly over the Ural Mountains, resulting in the reinforcement of this anomaly. In addition, from the lead-lag correlation analysis, we confirmed that the positive GPH anomaly over the Ural Mountains leads to a decrease in soil moisture anomaly in that region within approximately 2–5 days.Acknowledgements This work was supported by the National Research Foundation of Korea (NRF) grant funded by the government of Korea (MSIT; No. 2019R1A2C1008549).

The resolution of global climate models has increased dramatically in recent years. An increase in resolution can improve the numerical solution of the model, but at the same time, it carries an enormous increase in the cost of computations. This becomes more serious in the model development phase because numerous tests are required to assess the performance, and the performance should be assessed in a similar domain and resolution to the target simulations. We approach this problem by applying a vertical-meridional two-dimensional framework that can simulate many types of clouds and scale interactions. By neglecting the zonal extension, we can save computation time, and easily increase model resolution and the number of test cases. We developed the two-dimensional framework based on SAM (Khairoutdinov and Randall, 2003) following Satoh (1994), and have imported the new physics schemes planned to be used in a future global cloud resolving model version of DOE's Energy Exascale Earth System Model. We conducted a resolution sensitivity test for horizontal grid spacing of 16 km, 8 km, 4 km, and 2 km in a domain from the north-pole to the south-pole. For 4 km, 1000 days of simulation can be carried out with only 800 core-hours on a supercomputer. A large-scale circulation corresponding to the Hadley circulation in the real atmosphere is simulated. At higher resolution, the width of the simulated Hadley circulation tends to be wider and radiative cooling in the lower troposphere in the sub-tropical region becomes more marked. These suggest that the two-dimensional framework can serve as a testbed for assessing, amongst others, model parameterizations and their influence on the circulation. We expect this two-dimensional framework to accelerate the development of physics schemes in climate models.

Climate models and satellite remote sensing applications require accurate descriptions of ice cloud optical and radiative properties through parameterization of their scattering properties. While abundant irregularly shaped ice particle habits present a challenge for modelling ice clouds. An irregularly shaped ice particle habit (Voronoi model) has been developed and recently suggested to be effective in inferring the microphysical and radiative properties of ice clouds from satellite measurements. In this study, we develop a broadband ice cloud scheme based on the Voronoi model through parameterization for use in the Atmospheric Research Community Atmosphere Model Version 5 (CAM5). The new ice cloud scheme is compared with four ice cloud schemes (the Yi, Mitchell, Baum-Yang and Fu scheme), and is evaluated through the General circulation model version of the Rapid Radiative Transfer Model (RRTMG) and the top of atmosphere (TOA) shortwave (SW) and longwave (LW) cloud radiative forcing (CRFs) from CAM5. The Clouds and the Earth's Radiant Energy System (CERES) satellite data was selected as validation data. Results indicated that the Voronoi scheme can effectively reduce the TOA SW CRFs difference between CAM5 simulations and CERES data compared to the other four schemes, globally. Particularly for tropical ice cloud, the TOA SW CRFs of the Voronoi scheme is the closest to the CERES data. In general, it is found that the Voronoi model has ice cloud modelling capabilities similar to the default ice cloud scheme in CAM5.

The present study uses an object-based evaluation metric to examine the precipitation bias associated with the degree of convection organization over the tropics in the global storm-resolving models. The 40-day (1 August–10 September 2016) hindcast experiments of the DYnamics of the Atmospheric general circulation Modeled On Non-hydrostatic Domain (DYAMOND) intercomparison project are evaluated against the observational statistics derived from the high-resolution satellite rainfall products (GPM-IMERG and CMORPH). The hindcasts of the CWBGFS under the DYAMOND protocol are also included, which uses the unified parameterization of deep moist convection with the horizontal resolution of 15 km. The results show that most models simulate insufficient numbers of large object-based precipitation systems (OPS, > 300 km in scale), indicating weaker convection organization. The observation shows that the maximum precipitation within the object intensifies with increasing object size. All models capture this positive relationship but most of them overestimate the sensitivity. Although most models can reproduce the observed peak time of diurnal precipitation over land in the Maritime Continent, the simulated fractional contribution of different sizes of OPS to the total precipitation varies from model to model. In the observations, the primary precipitation contributor varies from the small OPS (< 100 km) to the mid-size OPS (100~300 km) as the precipitation increases through the diurnal cycle, while the fractional contribution by the large OPSs peaks at the midnight. Most models capture these observed features, but the models that explicitly resolve moist convection (ARPEGE-NH, ICON, and NICAM) significantly underestimate the overall contribution and the diurnal variation of the large OPSs.           

Entrainment is an important process affecting convective transport and is a key parameter in convection modeling. For over half a century, an inverse relationship between entrainment rate (and convective core size (R) has been used in modeling convective clouds. This study revisits the relation using multiple years of CloudSat observations. Results show that the inverse relationship between and R is robust. However, they are not scaled by a simple constant. Global satellite observations provide some guidance for choosing the range of the scaling parameter. Further, our analysis shows that continental convective clouds tend to have smaller λ and larger compared to oceanic convective clouds, suggesting that vertical mass transport by continental convection is more efficient. The satellite-based study of the relation has the potential to become a new diagnostic metric for assessing the representation of convection in weather and climate models. 

This study introduces a framework to evaluate the transition to aggregated convection under radiative-convective equilibrium simulations using the vector vorticity equation cloud-resolving model (VVM) coupled to a mixed-layer slab ocean. The framework introduces the competing effects between the convection-SST feedback and the convection-moisture feedback by modifying the initial SST gradient and the mixed layer depth. The convectional five-category scheme (VVM-Lin) and the predicted particle properties scheme (P3) are examined by the matrix formed by two factors. On the matrix, the boundary separating the aggregated and non-aggregated state shifts from 2K initial SST gradient in the VVM-Lin to 1K in the P3 when the mixed layer is 2m. It indicates that the P3 can generate stronger convection-moisture feedback even though the initial perturbation is weaker. When the initial SST gradient is 1.5K, the boundary moves from 2m mixed layer depth in the VVM-Lin to 1m in the P3. The results also point out that the convection-moisture feedback in the P3 can outweigh the stronger convection-SST feedback and lead to the convective aggregation. The isentropic analysis is applied to investigate the differences in the convective structures, and the results show that the stronger convective updrafts are responsible for the enhanced convection-moisture feedback in the P3. Besides, more high clouds in the P3 reduce the emission of the longwave radiation in the moist areas, which can accelerate the development of the aggregated convection. The proposed framework provides a uniform view of the process-based evaluation of convection aggregation among various cloud-resolving models that use different dynamics and physical parameterizations.

In a canonical view of the Madden–Julian oscillation (MJO), we recognize that MJO convection frequently starts to propagate from the Indian Ocean (IO). It is also a fact, however, that MJOs are sometimes initiated in the Maritime Continent (MC) and western Pacific (WP). Toward comprehensive understanding and better prediction of MJO initiation, we need to know how MJO initiation regions are determined. In this study, we examine what controls MJO initiation regions by focusing on MJOs categorized as those initiating in the IO, MC, and WP (refered to herein as IO-MJO, MC-MJO, and WP-MJO, respectively). We first analyze long-term observational datasets, and then we verify the observation-based hypothesis through a set of 15-year numerical experiments with a global nonhydrostatic MJO-permitting model (the Nonhydrostatic Icosahedral Atmospheric Model; NICAM) under a perpetual boreal-winter condition. The diversity of MJO initiation regions directly results from the modulation of areas where horizontal advective premoistening efficiently occurs via intraseasonal/synoptic-scale winds. This is supported by the difference in the zonal location of equatorial intraseasonal circulations established before MJO initiation, which is related to a spatial change in background convection forced by interannual SST variability. Compared to IO-MJOs (favored in the climatological background on average), MC-MJOs tend to be realized under the eastern-Pacific El Niño-like condition, as a result of eastward-shifted intraseasonal convection and circulation patterns induced by background suppressed convection in the eastern MC. WP-MJOs are frequently initiated under the central-Pacific El Niño-like and positive IO dipole-like conditions, in which the WP is selectively moistened with the aid of background enhanced (suppressed) convection over the WP (the southeastern IO and the central-to-eastern Pacific). These results indicate that the mutual relationship between the intraseasonal and interannual variabilities largely affects the charactetisics in MJO initiation.

A heavy rainfall event with the maximum accumulated rainfall >340 mm within 6h occurs over northwestern Taiwan coast on 17 May 2019 (LST). Before the rainfall occurs, the vicinity of Taiwan is under the unstable prefrontal southwesterly flow regime with a warm, moist tongue extending from the South China to Taiwan Strait. The marine boundary layer jet (MBLJ) transports moisture from low latitudes to Taiwan Strait in low levels.The Weather Research and Forecasting (WRF) model numerical simulations are used to analyze mechanisms for the heavy precipitation with terrain height sensitivity tests. During early morning, a distinct mesoscale convection system (MCS) initiates over northern Taiwan Strait as a result of the low-level convergence between the barrier jet (BJ) and weak westerly wind in mei-yu trough region. Then, the MCS keeps developing over the Taiwan Strait and extends eastward to northwestern Taiwan coast with the influences of the synoptic-related low-level jet (SLLJ) and strong westerly wind above. With the Taiwan terrain removed (increased), the BJ near northwestern Taiwan is not present (becomes stronger and wider). As a result, the low-level convergence and rainband weaken without Taiwan terrain and move slightly northward with the increased Taiwan terrain.

Recognizing needs to better utilize and improve monitoring of weather and climate extremes from space, the World Meteorological Organization (WMO) established a new flagship initiative - the Space-based Weather and Climate Extremes Monitoring (SWCEM). We started the SWCEM with the demonstration project for Asia-Pacific (2018-2019), and were able to bring clear benefits of translating science of satellite remote sensing to operational services at National Meteorological and Hydrological Services (NMHSs) in Member countries of WMO Regions II and V in a very short time. Recognizing SWCEM achievements in Asia and the Pacific, the Eighteenth World Meteorological Congress (Cg-18) in 2019 adopted the SWCEM Implementation Plan, endorsed its implementation from January 2020 in the region, and requested to consider the possibility of implementing  similar projects in Africa and South America. The demonstration project was focused on monitoring drought and heavy precipitation and it was implemented in geographical domain which covers the South-East Asia region and the Western Pacific Ocean area from 40°N to 45°S; 50°E to 120°W. The Japan Aerospace Exploration Agency (JAXA) and the Climate Prediction Center, National Oceanic and Atmospheric Administration (CPC/NOAA) provide satellite data and products for the region. SWCEM precipitation products produced by JAXA are based on the Global Satellite Mapping of Precipitation (GSMaP). CPC/NOAA provides SWCEM users with a similar set of products using the Climate Prediction Center morphing technique (CMORPH) satellite precipitation estimates. SWCEM space-based observations of precipitation have been incorporated into WMO activities strengthening capacity of Members, especially Small Island Developing States and Least Developed Countries, in climate change adaptation and disaster risk reduction. Satellite precipitation estimates and derived products are a significant contribution to strengthening Multi-Hazard Early Warning Systems. Currently, we are implementing it through the Climate Risk and Early Warning Systems (CREWS) projects.

A suite of observational global precipitation products are produced at NOAA Climate Prediction Center (CPC) for improved climate monitoring, climate diagnostics, and climate forecasts verifications. The product suite is composed of integrated satellite precipitation estimates, i.e. the CPC Morphing (CMORPH, Joyce et al. 2004; Xie et al., 2017) and the Gauge-CMORPH blended analysis (Wu and Xie 2016). The CMORPH precipitation products are utilized by the WMO Space based Weather and Climate Extreme Monitoring (SWCEM) program to help regional climate centers (RCC) and national meteorological and hydrometeorological servcies (NMHS) to monitoring and document extreme precipitation events over southeast asia and western Pacific regions. To facilitate the quantification of the extreme precipitation events, climatology of mean and extreme precipitation are defined using the CMORPH satellite precipitation data for a 20-year period from 1998 to 2017.  These include the mean, 99-, 95, 90- pecentile of daily precipitation as well as frequency of raining days for each 0.25^olat/lon grid and for each pentad period throughout the annual cycle.  Total precipitation, anomaly relative to mean climatology (normal), ratio of the 99-percentile, and standardized precipitation index (SPI) are computed to quantify the precipitation extremes.  These derived products are outputed in both digital and graphics formats and distributed to the participating members of the WMO program. Details of the CPC satellite precipitation will be reported at the AOGS meeting.

The second generation CMORPH (CMORPH2) has started real-time production of high-quality, high-resolution pole-to-pole global precipitation estimates since April 2017.  CMORPH2 is constructed in two steps. First, global fields of 30-minute precipitation, called CMORPH2_RAW, are constructed through integrating rainfall and snowfall rate retrievals from passive microwave (PMW) measurements aboard all available low earth orbit (LEO) satellites, precipitation estimates derived from infrared (IR) observations of geostationary (GEO) and LEO platforms, and model precipitation forecast from the NCEP operational global forecast system (GFS). A following step is applied to remove the bias in the CMORPH2_RAW through comparison against concurrent daily gauge analysis over land and calibration against GPCP monthly merged analysis V3.1 over ocean.  The final output, CMORPH2_ADJ, is generated on a 005^olat/lon grid over the globe (90^oS-90^oN), together with the fraction of solid precipitation computed from the surface air temperature with the model of Sims and Liu (2015).  A comprehensive examination has been conducted to examine the performance of the CMORPH2 real-time production in capturing precipitation and its temporal and spatial variations through comparison against CPC daily gauge analysis and radar estimates of precipitation over CONUS (MRMS), Alaska (M3QS), and Taiwan (CWB radar products). CMORPH2 near real-time precipitation estimates generated at various latency levels present very good quantitative consistencies, enabling quantitative applications (e.g. forcing forecast models) in a real-time setting.  The performance of the real-time produced precipitation estimates refines stably with the latency with the increased amount of input information, especially during the first 6 hours.  Substantial improvements are observed in the CMORPH2 in capturing and quantifying snowfall and cold season precipitation compared upon its predecessor CMORPH1. At the AOGS, progress on retrospective processing for the CMORPH2 will be also reported.

The WMO has been initiated the Space-based Weather and Climate Extremes Monitoring (SWCEM) with the recognizing that there is a need to better utilize and improve the monitoring of weather and climate extremes from space. The Japan Aerospace Exploration Agency (JAXA) has participated as one of meteorological satellite operators since the SWCEM Demonstration Project East Asia and Western Pacific regional subproject (SEMDP-EAWP) that began in 2018 with a duration of two years. JAXA has been developed the Global Satellite Mapping of Precipitation (GSMaP) products which have a resolution of 0.1 degrees and updated every hour, as part of the Global Precipitation Measurement mission, and provided this product for SEMDP-EAWP. To demonstrate a usefulness for precipitation extremes monitoring in EAWP region, case analysis and statistical analysis for both heavy rainfall and drought are investigated using Near-real-time Gauge-adjusted Rainfall Product (GSMaP_GNRT) that is the most suitable for monitoring. Heavy rainfall is here defined as when the percentile value calculated for the period from April 2000 to March 2019 exceeded the 90th percentile, and drought is defined by Standardized Precipitation Index (SPI) derived from GSMaP_GNRT since April 2000. These verifications were performed by comparing with Gauge-Based Analysis data provided by the NOAA CPC. For daily and weekly heavy rainfall, it was indicated that a threat score increases as total precipitation increases, and higher the percentiles of the threshold, the lower the detectability. Furthermore, it was suggested that a threat score varies greatly from region to region; Australia is relatively high, but Indonesia is low. For short-term drought from one-month to three-month, correlation coefficients in EAWP region are 0.712 in 1-month SPI and 0.735 in 3-month SPI, and the longer the accumulation period of precipitation, the higher the detection rate. Moreover, it seems that GSMaP_GNRT tends to detect drought intensity somewhat weakly.

The availability of an accurate rainfall dataset with good spatial coverage is important for many applications. For the purpose of climate monitoring and analysis, the availability of rainfall data product in the form of gridded dataset with long time series is necessary. Rainfall is traditionally measured with rain gauges; this type of in-situ measurements has the limitation of areal representation. In regions where the geographical conditions are challenging, such as high mountainous area, desert area, or dense forests, in-situ observations are often not available or only very sparsely distributed. In these types of area, to have a reliable rainfall estimate is difficult. Spaceborne rainfall observations from satellite remote sensing estimates offer an alternative estimation of rainfall, and provide its spatial distribution information.  Two satellite rainfall datasets, CMORPH and GSMaP, used in the Space-based Weather and Climate Extremes Monitoring (SWCEM) project are examples among a few others. Although satellite data products give better spatial information, they are subject to bias. For operational purpose, the Agency for Meteorology Climatology and Geophysics (BMKG) has developed a blended rainfall dataset combining in-situ observation with satellite estimates, to obtain a dataset with benefits of the two sources of information. The blended dataset is constructed using a simple geostatistical approach, where the satellite data is used as a secondary drift variable informing the spatial structure, and the in-situ value data is retained after interpolation. In this presentation, examples of blended products produced by BMKG are presented. It is demonstrated that the blended interpolation captured better the spatial distribution of the rainfall, while the interpolation retains the original observation values in the in-situ measurement locations. This blending procedure using GSMaP has been used operationally for rainfall monitoring and analysis, and the dataset was constructed since the year 2000 following the availability of GSMaP. 

Satellite rainfall estimation has received much attention with its increasing reliability matched with extensive coverage and high update frequency. WMO’s project Satellite Weather and Climate Extreme Monitoring for East Asia and West Pacific (SWCEM-EAWP) explores the usage of satellite rainfall estimates for extreme weather and climate purposes in the region. However, the performance of the rainfall estimates may vary seasonally as changing rainfall systems give rise to different challenges in rainfall monitoring.Using the high-density rain-gauge network in Singapore, the performance of SWCEM’s satellite rainfall estimates products from the partnering service providers, JAXA (GSMaP NRTv6) and NOAA CPC (CMORPH CRT), is evaluated. Further calibration using quantile mapping with gamma distribution is subsequently applied at different timescales, with promising results that support the usefulness of such calibration techniques for performance enhancement of the satellite rainfall estimations.*JAXA: Japan Aerospace Exploration Agency
NOAA CPC: National Oceanic and Atmospheric Administration Climate Prediction Center

Global Satellite Mapping of Precipitation (GSMaP) version 6 (GSMaP_GNRT6) is validated using rain gauge data over four contrasting topographic sub-regions of Peninsular Malaysia i.e., west coast (WC), foothills of Titiwangsa range (FT), inland-valley (IN), and east coast (EC) sub-regions. Result shows that the GSMaP_GNRT6 perform best over coastal sub-regions (WC and EC sub-regions), underestimates over FT and overestimates in IN sub-regions. GSMaP_GNRT6 also tends to perform better for moderate rainfall compared to heavy rainfall events. For operational purpose, the Malaysian Meteorological Department has developed a blended daily rainfall dataset combining of 160 rain gauge stations with GSMaP_GNRT6. The blended dataset is constructed using Barnes successive correction approach, where the GSMaP_GNRT6 is taken as an initial guess field and rain gauge as the ground truth. Overall, the blended dataset captured better spatial rainfall distribution over the Peninsular Malaysia. Verification work of blended dataset using the leave-one-station-out cross validation approach is currently going on.

Effective mitigation of surface ozone pollution entails detailed knowledge of the contributing precursors’ sources. We use the GEOS-Chem adjoint model to analyze the precursors contributing to surface ozone in the Beijing-Tianjin-Hebei area (BTH) of China on days of different ozone pollution severity in June 2019. We find that BTH ozone on heavily polluted days is sensitive to local emissions, as well as to precursors emitted from the provinces south of BTH (Shandong, Henan, and Jiangsu, collectively the SHJ area). Heavy ozone pollution in BTH can be mitigated effectively by reducing NO_x (from industrial processes and transportation), ≥C₃ alkenes (from on-road gasoline vehicles and industrial processes), and xylenes (from paint use) emitted from both BTH and SHJ, as well as by reducing CO (from industrial processes, transportation, and power generation) and ≥C₄ alkanes (from industrial processes, paint and solvent use, and on-road gasoline vehicles) emissions from SHJ. In addition, reduction of xylenes and ≥C₃ alkenes emissions within BTH would effectively decrease the number of BTH ozone-exceedance days. Our analysis pinpointed the key areas and activities for local and regionally-coordinated emission control efforts to improve surface ozone air quality in BTH. 

We developed incremental reactivity (IR) scales for 116 volatile organic compounds (VOCs) in a Chinese megacity (Guangzhou) and elucidated their application in calculating the ozone (O₃) formation potential (OFP) in China. Two sets of model inputs (emission-based and observation-based) were designed to localize the IR scales in Guangzhou using the Master Chemical Mechanism (MCM) box model, and were also compared with those of the U.S. The two inputs differed in how primary pollutant inputs in the model were derived, with one based on emission data and the other based on observed pollutant levels, but the maximum incremental reactivity (MIR) scales derived from them were fairly similar. The IR scales showed a strong dependence on the chemical mechanism (MCM vs. Statewide Air Pollution Research Center), but the discrepancy between China and the U.S. using a similar chemical mechanism was not large. With a given chemical mechanism, the MIR scale for most VOCs showed a relatively small dependence on environmental conditions. However, when the NOx availability decreased, the IR scales became more sensitive to environmental conditions and the discrepancy between the IR scales obtained from emission-based and observation-based inputs increased, thereby implying the necessity to localize IR scales over mixed-limited or NOx-limited areas. This study provides recommendations for the application of IR scales, which has great significance for VOC control in China and other countries suffering from serious O₃ air pollution.

East Asia is a region of increasing economic growth which has led to severe pollution in urban areas. During the NASA KORUS-AQ field campaign in Seoul, South Korea, a period of haze was observed where PM_2.5 rapidly increased to the highest levels observed during the campaign and exceeded local air quality standards. Observational evidence attributed this increase to enhanced local production of secondary inorganic aerosol (SIA). However, models have difficulty reproducing PM_2.5 levels across East Asia, particularly the composition of SIA (sulfate, nitrate, and associated ammonium). Models severely underestimate sulfate during haze and generally have poorly constrained budgets for nitrate production, sometimes resulting in compensating biases or large overestimates that impact aerosol radiative properties and the simulation of AOD. These biases hinder the ability of models to assess the impact of emission changes on PM_2.5levels or determine the amount of pollution resulting from transboundary transport. Here, we use the GEOS-Chem chemical transport model to interpret observations from KORUS-AQ and improve the model ability to reproduce sulfate levels during haze and explore potential causes of biased model nitrate, a long-standing problem in models in East Asia as well as in the United States. We find that long-term ground-based observations combined with detailed observations of atmospheric composition from aircraft provide an unprecedented dataset to improve our understanding of SIA and PM_2.5.

The worldwide expansion of urbanization creates higher population densities in complicated urban environments with dense road networks and complicated built-environments. A combination of disproportionate emissions and modified (micro-)meteorology in urban canopy lead to spatiotemporal heterogeneity of air pollution, which then, affects pedestrian and traffic users’ exposure to air pollution because the relatively short-period exposure to highly elevated pollution near roadways can contribute to a significant fraction of daily exposure. Recently, a mobile platform and cost-effective sensors have been widely used in air quality studies to investigate heterogeneous air pollution characteristics in urban areas as well as to estimate vehicular emissions. This presentation summarizes the preliminary results from recent field intensives using a mobile platform and a cost-effective sensor network in various micro-built-environments. We briefly introduce the field intensives we have recently been conducting and present the preliminary results from the experiments. The focuses of the studies include quantification of specified VOCs emission factors from vehicles in tunnels; emissions and evolution of vehicle-emitting aerosols as a traffic plume transports horizontally and vertically; spatially dense air quality monitoring with a sensor network in an intra-community scale; micro-meteorological effects on air pollutant levels in different micro-built-environments. More details concerning these preliminary findings are presented in this presentation and several companion posters. 

The Tibetan Plateau (TP), also known as the 'Third Pole', is the highest and most widespread highland in the world. In the past few decades, the TP has been greatly affected by global climate change. However, the observation data over the TP is very limited, and assessments of the impact of climate change on air quality are hindered. In this study, we simulate the concentration changes of major air pollutants in the TP and the surrounding regions in January, April, July and October in 2019 using the Community Multi-scale Air Quality (CMAQ) model at the resolution of 36 × 36 km². The meteorological conditions are generated by the Weather Research and Forecasting (WRF) model. The simulated surface concentration shows seasonal differences. Especially, large variability of the carbonaceous components was observed due to the change of meteorological conditions. The study reveals the importance of controlling emissions in the TP in different seasons and may provide a reference for designing effective control strategies.

The Yangtze River Delta (YRD) urban agglomeration is one of the most developed regions in China. During recent decades, this region has experienced severe regional haze and photochemical smog pollution problems. In this study, we used a source-oriented chemical transport model to quantitatively estimate the effects of inter-city transport on fine particulate matter (PM2.5) and ozone (O3) among the 41 cities in the YRD urban agglomeration during the EXPLORE-YRD (EXPeriment on the eLucidation of the atmospheric Oxidation capacity and aerosol foRmation, and their Effects in the Yangtze River Delta) campaign (May 17 to June 17, 2018). The results show that inter-city transport is very significant in the YRD region. On average, the emissions from the local city, the other YRD cities, and the regions outside of the YRD contribute 25.3%, 49.9%, and 24.8% to the PM2.5, respectively, and they contribute 33.7%, 46.8%, and 19.5% of the non-background O3, respectively. On PM2.5 or O3 pollution days, the transport contribution from the nearby cities becomes much more important, while the local emissions and the transport from farther away cities become less important. The results also suggest that the cities within a distance of 184 km and 94 km contribute 60% of the PM2.5 and O3, respectively. Therefore, we recommend that regional cooperative control programs in the YRD consider emission controls over cities within these ranges. The range for primary PM2.5 (92 km) is very different from that for secondary PM2.5 (515 km). Cooperative emission controls of SO2 and NOx on a much larger regional scale are required to reduce the secondary PM2.5 in the YRD.

Air pollution is one of the important problems in Korea, which leads to causes of mortality. To assess the air quality and spatio-temporal characteristics of the atmospheric particulate matter (PM), a detailed analysis is performed over South Korea using 2015-2010 data. The meteorological conditions play a significant role in the formation, transportation, and deposition of air pollutants. This study investigates the relationship between meteorology and air pollutants concentration using 220 station datasets. The monthly PM signatures show maximum PM_2.5 in January and minimum in July; PM₁₀ was high and low in April and July, respectively. The clear seasonal trends were also detected, the PM_2.5 concentration was high during the winter followed by spring, autumn, and summer; PM₁₀ was high in spring, then winter, autumn, and summer. High PM_2.5 was observed over Chungcheongbuk and Jeollabuk provinces, then Gyeonggi, Seoul, and Incheon. While, PM₁₀ was high over Gyeonggi, followed by Seoul, Incheon, Chungcheongbuk, and Jeollabuk. Further, Ulsan is an industrial powerhouse of Korea, thus PM₁₀ was high over this region. Additionally, high values of CO, Black Carbon, SO₂, and SO₄ were also noticed in the capitol and surrounding regions of Gyeonggi, Incheon, Chungcheongbuk, and Jeollabuk, and Ulsan provinces due to heavy traffic and emissions of industries. We also explored the effects of meteorological factors on PM concentration. The high air pressure and downdraft accumulate PM concentration. The precipitation reduces the PM caused by wet removal and washing effect. The convection and evaporation loss processes remove the pollutants in high temperature and relative humidity environment. The high radiation/ground heating promotes the upward movement and diffusion of PM. The wind speed accumulates PM may cause by Korea occupied by mountainous.  The Korean government implementing strict air pollution control measures, thus the decreasing trend of PM has been observed from 2015 to 2020.

The Geostationary Environment Monitoring Spectrometer (GEMS) was launched in February 18, 2020 to monitor air quality over Asia at an unprecedented spatial and temporal resolution from a geostationary Earth orbit (GEO) for the first time. After arriving at GEO in March, GEMS started the first observation in April, 2020 and finished in orbit tests successfully. With the UV–visible spectrometers at sub-nm spectral resolution and sophisticated retrieval algorithms, hourly estimates of the column amounts of atmospheric pollutants (O3, NO2, SO2, HCHO, CHOCHO, and aerosols) are obtained. To date, all the UV–visible satellite missions monitoring air quality have been in low Earth orbit (LEO), allowing one to two observations per day. With UV–visible instruments on GEO platforms, the diurnal variations of these pollutants can now be determined. Details of the GEMS mission status are presented, including instrumentation, scientific algorithms, and initial results. GEMS onboard the Geostationary Korea Multi-Purpose Satellite 2 (GEO-KOMPSAT-2) satellite series, also hosts the Advanced Meteorological Imager (AMI) and Geostationary Ocean Color Imager 2 (GOCI-2). These three instruments will provide synergistic science products to better understand air quality, meteorology, the long-range transport of air pollutants, emission source distributions, and chemical processes. Faster sampling rates at higher spatial resolution will increase the probability of finding cloud-free pixels, leading to more observations of aerosols and trace gases than is possible from LEO. GEMS will be joined by NASA’s Tropospheric Emissions: Monitoring of Pollution (TEMPO) and ESA’s Sentinel-4 to form a GEO AQ satellite constellation in early 2020s, coordinated by the Committee on Earth Observation Satellites (CEOS).

The recent launch of the Geostationary Environment Monitoring Spectrometer (GEMS) provides an important catalyst for increased dialogue and cooperation among Asian countries to address air quality. Other emerging efforts to support GEMS validation and international cooperation include the Pandora Asia Network (PAN) and the Pan Asia Partnership for Geospatial Air Pollution Information (PAPGAPI). These efforts represent long-term commitments to bridging satellite observations with ground-based monitoring to inform air quality. Aircraft observations can provide invaluable context to the satellite and ground-based perspectives that are used more routinely to inform air quality models used for both forecasting and attribution. Important information from aircraft includes measuring detailed composition for source fingerprinting, vertical profiling of composition for satellite validation and model assessment, observing chemical and dynamical processes affecting secondary pollution (i.e., fine particles and ozone), relating specific VOC mixtures to satellite HCHO, providing fine scale pollution mapping with remote sensors, etc. Such information is critical for understanding the local factors influencing air quality for a specific location, quantifying emission sources, and assessing potential mitigation strategies for decision makers. ASIA-AQ proposes to provide airborne observations over three to five Asian megacities with repetitive observations that will observe the diurnal and vertical distribution of primary emissions and secondary pollutants with at least four flights over each location. In combination with satellite and ground observations, data would support analyses for assessment of emissions, model evaluation, process-level understanding of secondary pollutants (i.e., fine particles and ozone), and satellite validation and interpretation. Current status of the ASIA-AQ white paper, nominal plans, and opportunities for involvement will be presented.

Chinese Gao Fen-5 (GF-5) is a new satellite, which was launched in May, 2018. Directional polarimetric camera (DPC) is a polarimetric sensor onboard on GF-5 satellite, aimed to monitor aerosol accurately. DPC can measure surface object in multi angle (max to 12) in three polarimetric bands (490 nm, 670 nm and 865 nm), and the spatial resolution is about 3.5 km. In the paper, we applied DPC data to monitor PM_2.5. Firstly, from multi-angle polarimetric DPC data in 670 nm and 865 nm, fine-mode aerosol was retrieved by best-fitting method. Secondly, assisted by the weather and ground observation data, geographically weighted regression (GWR) method was used to retrieve PM_2.5 from DPC aerosol. With DPC data over Beijing-tianjin-hebei Areas from January to March, 2020, our algorithm was applied to PM_2.5 concentration retrieval. The retrieved PM_2.5 images show that our algorithm can reveal the spatial distribution of PM_2.5, and the validation with the ground-based sites showed that the correlation was greater than 0.8 and the relative error was about 10 μg/m³. 

Researches of air quality over coastal regions require high-resolution Aerosol Optical Depth (AOD) with high accuracy over coastal waters, while they are unavailable in the MODIS Dark Target (DT) product. The high water leaving radiance at visible and near-infrared bands over shallow and turbid coastal waters violate the dark surface assumption in the DT algorithm, hence these pixels are masked in the operational product. Moreover, the coastal shallow and turbid waters masking approach may fail sometimes, thus the algorithm DT could be wrongly applied to these pixels, which leads to abnormal high AOD retrievals as the high water leaving radiance will be mis-interpreted as the elevated aerosol loadings. Additionally challenges includes how to avoid recognizing land or land-contaminated pixels as waters and classifying the clear-sky pixels over coastal waters as clouds. This study improves on our previous 10-km-resolution coastal waters AOD retrieval algorithm in which the MODIS 2.1 µm Top of Atmosphere (TOA) reflectance was used to retrieve AOD over turbid coastal water because water-leaving radiances at this band is negligible regardless of surface water turbidity. The main improvements consist of (1) use 2.1 µm TOA reflectance to conduct smoothness test for cloud detection, which could additionally detect land and land-contaminated pixels and retain clear-sky pixels over turbid waters; (2) use auxiliary dataset to label constant shallow waters to avoid the misuse of DT algorithm; (3) retrieval AOD at the resolution of 1 km rather than 10 km. The 1-km coastal water AOD retrievals are well validated with measurements from Maritime Aerosol Network (MAN). Additionally, yearly mean operational AOD has larger values over shallow and turbid coastal waters than adjacent land, as DT algorithm cloud be wrongly applied sometimes, while this study retrieves reasonable AOD values and have more retrievals over shallow and turbid coastal waters.

Designed as the successor of Environment-1 (HJ-1) A/B satellites in Chinese Environment and Disaster Monitoring and Prediction Satellite Constellation, Environment-2 (HJ-2) A/B satellites have been launched in 2020, which can provide better remote sensing performance, image quality, imaging performance and satellite reliability. A new space-borne instrument called Polarized Scanning Atmospheric Corrector (PSAC) also board on HJ-2 satellites, aiming to provide the atmospheric properties for synchronous atmospheric correction of the main sensors, such as the charge-coupled device (CCD) cameras onboard the same satellite. PSAC was designed with multispectral single-viewing polarimetric measurements from visible to shortwave-infrared wavelength. Based on the optimal estimation (OE) theory, we develop an optimized inversion algorithm for remote sensing of aerosol parameters from single-viewing intensity and polarization together over land, by taking full advantage of available multispectral polarimetric measurement information form the visible to shortwave infrared wavelength with the synthetic measurements of PSAC. In the theoretical framework for aerosol retrieval, the principal component analysis (PCA) method is used to reconstruct the multispectral surface reflectance by about three PCs extracted from the spectral library, and thus we can try to retrieve the weighting coefficient of each PC instead of the direct retrieval of surface reflectance. Meanwhile, a bidirectional polarized reflectance distribution function (BPDF) integrated in Unified Linearized Vector Radiative Transfer Model (UNL-VRTM) is used to describe the surface polarized contribution for the forward model and the retrieval test with synthetic and real data. Preliminary retrieval tests and results are also discussed.

To better understand the role of aerosols in climate change and their direct effects on human health, atmospheric aerosol properties have been monitored by various satellite sensors. The Geostationary Environmental Monitoring Spectrometer (GEMS), onboard Geokompsat-2B (GK-2B) satellites, is the first air quality monitoring sensor in geostationary earth orbit and was successfully launched on February 19, 2020. GEMS measures the hyperspectral radiances with 0.6 nm spectral resolution in UV and visible range over Asia during the daytime to provide air quality information. This study shows the first atmospheric aerosol monitoring results from GEMS after the launch of GK-2B satellites. The aerosol retrieval algorithm for GEMS uses 6 channels in UV and visible wavelength which has the advantage of measuring aerosol absorption and height information. The aerosol retrieval algorithm for GEMS is based on an optimal estimation method finding the optimized values for the aerosol optical depth (AOD), single scattering albedo (SSA), and aerosol loading height (ALH) to minimized differences between simulated radiance and observations. We present early GEMS aerosol products to discuss high aerosol loading cases over East Asia and analysis results as a case study. Our preliminary results show that the GEMS aerosol products have the advantages to capture the diurnal variation with high spatio-temporal resolution. Qualitatively good agreements and fine-scale features are shown in this case study. Further, the first validation of GEMS aerosol products is presented as compared to ground-based AERONET and other aerosol products obtained from the Advanced Himawari Imager (AHI) onboard Himawari-8.

The atmospheric electric field (AEF) measurement has been carried out by the Japan Meteorological Agency (JMA) at Kakioka geomagnetic observatory since 1929. The JMA terminated the AEF measurement the end of February 2021. Since many researchers have used the AEF data at Kakioka, several research groups such as universities examined whether or not successive observations could be conducted at Kakioka. In our grope, we installed one field mill with a broad range for the AEF intensity developed by Otowa Co. Ltd., sensitive and insensitive field mills developed by Boltek Co. Ltd. and two all-sky cameras at 18 of Feb., 2021. This observation was available on not only traditional fair-weather study but also the study of severe weather accompanying heavy lightning activities. The sampling of field mills are 20 Hz (Otowa) and 10 Hz (Boltek). The time stamp was synchronized by GPS. This paper is a preliminary report of the AEF measurement in the vicinity of the current AEF observation operated by our group. In the presentation, parallel measurements using JMA’s water dropper and our field mills are shown. In addition, a future plan of long-term AEF measurements at Kakioka is discussed.

Lightning observations provide important context to thunderstorm research, offering information on the amount and type of lightning that can be used as markers of convective strength and of severe weather potential. As part of an effort to study high impact storms, lightning observations from the University of Colorado Boulder deployed Low-Frequency (LF) radio Lightning Locating System (LLS) were collected during the RELAMPAGO field campaign in central Argentina (November 1 to December 15 2018), where some of the strongest storms on earth occur. Lightning type classification, unique to LF observations, may be used to quantify the prevalence of lightning types and their polarity, such as compact intra-cloud lightning discharges (CIDs) and energetic intra-cloud pulses (EIPs), associated with different storm types, particularly severe storms. To investigate this, first lightning characteristics of different storms need to be studied. Further investigation on how their temporal evolution relates to storm microphysical and kinematic processes may help understand their production.
LF data products were derived from lightning observations by an array of four LF radio receivers deployed in RELAMPAGO. The LF Level 1 data product contains radio waveforms observed from lightning emissions, commonly referred to as 'sferics'. The LF Level 2 data product provides time, location and peak current estimates for lightning events and flashes. Lightning type classification can be performed on the Level 2 events and corresponding Level 1 sferics based on a few waveform shape parameters. In this study, we investigate the temporal evolution of different lightning types for different thunderstorm types during RELAMPAGO, focusing primarily on measures of total lightning, polarity, and lightning type. A few case-studies of storms that produced severe weather are also investigated by comparing the lightning evolution with the occurrence of hail and if available other storm kinematic and microphysical information.

Lightning density variation was analyzed in the area with borders of 56-72 N, 105-150 E, which was divided into 12 parts of 4x15 degrees in latitude and longitude. The spatial distribution of lightning density was obtained from the data (2009-2019) of World wide lightning location network (WWLLN), and lightning pulse number estimates were done by the data of two single-point lightning detectors: Stormtracker (Boltek Corporation) (effective radius of 480 km, 2009-2019) and lightning direction finder (effective radius of 1200 km, 1999-2016) of our construction. The lightning density showed some relation to orography and a linear decline with increasing latitude. Intense thunderstorms in the southern and central parts of Yakutia (up to 68 N) were often formed under the influence of cyclonic activity in the southwestern and southern regions of Yakutia. In the years with a peak lightning number in the north of Yakutia, lightning activity in southern parts was relatively low until 2018. Annual lightning variation in the eastern parts was not related to annual variation in the central parts of Yakutia due to the obstructive Verkhoyansk ridge. The sharply continental climate of Yakutia in the summer season is characterized by meridional outlets of southern cyclones from Transbaikalia, occurring during the disturbance of the zonal transport dominant in the winter season in Yakutia. Southern outlets produce intense summer thunderstorms, that resulted a significant contribution to the total lightning number per season. In the rear of the southern cyclones, the Arctic air inflows caused periods of significantly reduced lightning activity before or after peaks in the seasonal variation of lightning number. This work was supported by the Ministry of Science and Higher Education of the Russian Federation and Siberian Branch of the Russian Academy of Sciences (registration number АААА-А21-121012000007-4).

This study verifies the performance of WRF model to simulate strong hailstorm over Bangladesh which occurs several times on 27^th February 2019. The simulation techniques with (DA) and without data assimilation (CNTL) both are used in this study. The model simulation is run with 02 nested domain of 9 km and 3 km horizontal resolution using NCEP Global Forecast System (GFS) data from 0000 UTC of 27^th February 2019 as initial and lateral boundary conditions. Firstly, the model simulated environmental parameters are varied with the BMD observed data for calibration of the microphysical schemes; thereafter the calibrated model is used to observe the hailstorm event. But in the control run of the WRF model, it cannot capture the exact location (data has been collected from National Dailies), intensity of hailstorm over Bangladesh. Advanced Microwave Sounding Unit (AMSU) data is assimilated in 3DVar with the first guess of WRF model for data assimilation (DA) run to gain better initial conditions in the performance of model. Then outputs are analyzed root mean square error are calculated for both CNTL and DA to compare and/or evaluate the model performance. The results from the study suggest that model simulation with assimilation perform better to find the location, precipitation, pressure, temperature, accumulated rainfall, and CAPE value than the simulation without assimilation. The model simulated weather elements and microphysical parameters have been found closely comparing with the BMD observation during the hail event. Therefore, this kind of study can directly help to understand the atmospheric conditions more precisely to forecast the hailstorm over Bangladesh and operational hailstorm forecast may be improved through satellite radiance data assimilation to reduce the casualties and properties damages.

Atmospheric stability is a measure of the atmosphere's tendency to discourage vertical motion which is directly correlated to different types of weather systems and their severity. In this study an attempt has been made to inspect the evaluating of thunderstorm over Dhaka (31^st March, 2019) using Advanced Weather Research and Forecasting (WRF-ARW) Model. The model is run for 72 hours with 03 nested domain of 09 km,  03 km and 01 km horizontal resolutions using 0.25º X 0.25º six hourly global data assimilation system (GDAS) data from 0000 UTC of one day before of starting date to one day after of the required events as initial and lateral boundary conditions. The model performance is evaluated by calculating hourly instability indices (VTI, TTI, KI, CTI, MCAPE, MCIN, BRN, LI, SI, SWI) value and compared with the threshold value of indices. Different meteorological parameters such as mean sea-level pressure, temperature, winds at upper (300 hPa) and lower (925 hPa) levels, relative humidity along with vertical cross section are also studied by the model and compared with the favourable conditions for development of TS. Area average rainfall (hourly) value is also calculated and compared with indices value to comprehend the nature of thunderstorm. Observing the indices value it is seen that all indices value increase sharply 5-6 hours before of TS occurring and MCAPE is giving more reliable result.  Moreover, this study shows that inner two domains (3 and 1 km resolution, giving almost same values) are giving better results than outer one and which indices are more probable in forecasting of TS for our country as well as giving less Root Mean square Error. From the simulated and validated results it can be concluded that the model performance of instability indices can be used as forecasting of TS over Bangladesh.

Recently, the studies of discharge phenomena in the atmosphere have improved our current understanding of electric charge associated effects on the lithosphere-atmosphere-ionosphere coupling (LAIC) processes. The experimental evidence of existing stressed-rock currents supports our speculation that the electric coupling accompanying pre-earthquake activity may happen. These phenomena associated with earthquake precursors were known, such as atmospheric conductivity anomalies and total electron content variations in the ionosphere. The LAIC modeling includes the radon ionization leading to charged aerosols and ionosphere dynamics with an imposed zonal electric field or currents. Except those anomalies associated with precursors activities, we propose an acoustic wave generation mechanism which may explain the observed acoustic wave of earthquake precursors. As rocks are subject to stress, surface charges can be induced by dynamo current near ground. Pressure gradient may be formed by balancing the repulsive force of established surface charges. As the dynamo current-driven surface electric field suddenly stops, the surface charge region cannot be sustained. The pressure over the pre-exiting surface charge region may be increased and cause the acoustic wave. We simulated the generated acoustic waves in the atmosphere. The generated acoustic waves not only transport momentum/energy to higher altitude in the atmosphere, but also may modulate the airglow intensity in the mesosphere.